Lesson 1 · Free · 2 min
$ git worktree add .claude/worktrees/discount -b worktree-discount
Watch lesson 1 freeThe loop behind every agent · 2 min

Engineering with Coding Agents

AI Engineering · 27 lessons · 59 min · Intermediate

A working engineer's course on using coding agents such as Claude Code, Codex CLI, Copilot's agent mode and Cursor without lowering your bar. You learn how an agent really works, how to give it context, how to delegate in reviewable pieces, how to verify and secure what it writes, and how to run all of it as a team.

What you'll be able to do

  • Explain the agent loop and predict what fills the context window and when it fails
  • Write a short instruction file and a spec an agent can build against
  • Split work into reviewable pieces and run parallel agents in git worktrees
  • Read an agent's diff for weakened tests, invented APIs and scope creep
  • Lock down permissions, sandboxes, secrets and dependencies around an agent
  • Set a review policy, CI gates and honest metrics for a team using agents

How the lessons work

Lessons that move

Narrated, animated lessons with real terminals and code on screen.

Pause and ask

Stuck? Ask out loud or type. The answer plays as a short video made for your question.

Check yourself

Short quizzes inside the lessons show whether it stuck.

Course content

27 lessons · 59 min · 2 lessons are free
Unit 1 · How an agent works3 lessons · 7 min
01The loop behind every agent2 minFree02The context window and what fills it2 minFree
03Meet Ledgerline, the project you will extend2 min
Unit 2 · Giving it context4 lessons · 8 min
04Instruction files: CLAUDE.md, AGENTS.md and friends2 min
05Keep instructions short, enforce the rest2 min
06Write a spec the agent can build against2 min
07Examples, pointers and skills2 min
Unit 3 · Decompose and delegate5 lessons · 11 min
08Size the task to one reviewable diff2 min
09Plan first, then build2 min
10Sub-agents and their own context2 min
11Parallel agents with git worktrees2 min
12When the agent is the wrong tool2 min
Unit 4 · Verify what it produces4 lessons · 9 min
13Tests first, written by you2 min
14How to read an agent's diff2 min
15Mistakes to look for in agent code3 min
16Evals for your own prompts2 min
Unit 5 · Guardrails5 lessons · 11 min
17Permissions and modes2 min
18Sandboxes: limits the OS enforces2 min
19Hooks for rules with no exceptions2 min
20Secrets and prompt injection2 min
21Packages the agent invents2 min
Unit 6 · Team practice4 lessons · 9 min
22Review policy and ownership2 min
23CI around the agent2 min
24Measuring what matters3 min
25Cost and onboarding2 min
Unit 7 · Capstone2 lessons · 4 min
26Capstone: ship the refund feature2 min
27Capstone: the failure and its diagnosis2 min